This study presents a novel interdisciplinary framework for modeling the dynamics of emotionally charged information dissemination in social networks, incorporating principles inspired by quantum theory. We developed a multi-agent system to model emotional interactions in social networks, using valence–arousal dynamics and quantum-inspired principles. The model captures key features of online emotional communication, such as indirect feedback, asynchronous expression, and nonlinear amplification effects. Emotional states are defined as internal variables that evolve in response to both deterministic social stimuli and stochastic cognitive fluctuations. Sentiment analysis was conducted using a state-of-the-art large language model (GPT-4 Omni) on over 50,326 user comments and 2,246 posts from a public Russian social network, VK. Empirical findings reveal structured emotional patterns and interference-like phenomena, akin to quantum interference and electromagnetically induced transparency. Principal component analysis confirmed that emotional tone and engagement are orthogonal dimensions, supporting the analogy between psychological states and quantum observables. These results support the hypothesis that emotionally mediated digital communication may exhibit quantum-like characteristics, offering a promising alternative to classical probabilistic approaches. The proposed model has practical implications for early detection of emotional polarization, information spread forecasting, and enhancing digital resilience against disinformation.